Abstract
Using Y145F and Y112/128F SLP-76 mutant KI mice we show that altering proximal TCR signaling can differentially affect T cell fate choices independent of the external inflammatory environment. In vitro T helper polarization studies reveal defects in Th2 and Th17 but not Th1 differentiation in the absence of SLP-76 tyrosines. Conventionally induced Th2 cells are dependent on SLP-76 tyrosines for IL4 production but not
competence. Induced Th17 cells, on the other hand, are dependent on SLP-76 tyrosine signals for both IL17a production and Il17a competence. Furthermore, these studies have uncovered a previously undefined subset of CD44hi CD4+ T cells that acquire the potential to produce IL17a in vivo, independent of SLP-76 tyrosines. To define the effects of dampened TCR signals on CD8+ T cell effector and memory differentiation, we infected WT and SLP-76 KI mice with the Armstrong strain of Lymphocytic
Choriomeningitis Virus (LCMV). Data from these studies support a model in which
altered TCR signals can determine the rate of memory versus effector cell differentiation, independent of initial T cell expansion. Furthermore, we show that TCR signals sufficient to promote CD8+ T cell differentiation are different than those required to elicit
inflammatory cytokine production, reminiscent of our observations in the in vitro Th2 polarization studies.
Introduction
The heterogeneity of possible T cell effector responses reflects the heterogeneity of possible offending pathogens and is essential for an effective adaptive immune system. Integration of multiple extracellular signals including those from cytokines, pathogen associated molecular patterns, co-stimulatory receptors and peptide:MHC complexes is required for the establishment of a T cell response that is specific to the nature of the inciting pathogen. Naïve CD4+ T cells can differentiate into various T helper (Th)
lineages including Th1, Th2 or Th17 effector cells as well as regulatory T cells, follicular helper T cells, memory cells and other less well-defined cell types (Zhu et al., 2010). Naïve CD8+ T cells differentiate into effector cytotoxic T lymphocytes (CTLs) and they either undergo terminal effector differentiation or they acquire a memory phenotype (Kaech and Wherry, 2007). While many studies have focused on the role of cytokines for T cell effector choices and function, the role of TCR signals on these processes is still not fully understood.
It is apparent that altering distinct TCR signals can have a profound effect on the effector functions of T cells. For example, mutations in the TCR proximal signaling adaptor LAT result in aberrant cytokine production and lymphoproliferative disease (Sommers et al., 2002). Furthermore, studies using altered peptide ligands (APLs) have shown both in vitro and in vivo that the strength of the interaction between the TCR and pMHC can influence lineage choices. In CD4+ T cells, these studies focused on Th2 versus Th1 differentiation and showed that, in the absence of exogenous cytokines, strong interactions induce higher levels of ERK phosphorylation which correlates with inhibition of Th2 differentiation and acquisition of Th1 effector function (Jorritsma et al., 2003; Kumar et al., 1995; Pfeiffer et al., 1995; Tao et al., 1997). The role of TCR signal
strength in the presence of Th polarizing cytokines has not been established. However, it is established that, downstream of the TCR, translocation and activation of NFAT and NFκB are important for inducing the distinct patterns of gene expression that define Th
cell effector lineages (Hermann-Kleiter and Baier, 2010; Zhu et al., 2010). Activity of NFAT and NFκB are dependent on TCR-induced Ca2+ mobilization and DAG signals
respectively, as well on their transcriptional binding partners that are differentially expressed in response to distinct cytokine signals. Following differentiation, bona fide Th2 and Th17 cells, in particular, require further TCR-induced NFAT transcriptional activity to produce their signature cytokines (Gomez-Rodriguez et al., 2009; Hermann- Kleiter and Baier, 2010; Kosaka et al., 2006).
CD8+ T cell differentiation and function has largely been tested with in vivo viral and bacterial infection models, which have revealed a typical pattern of the CD8+
response that begins with a robust expansion phase and, following antigen clearance, a contraction phase and finally a memory maintenance phase. Naïve CD8 + T cells acquire effector function upon infection. The pool of these effector cells is composed of terminally differentiated short-lived effector cells (SLECs) and memory precursor effector cells (MPECs). Following antigen clearance the SLECs die by apoptosis and a subset of the MPECs survive and become long-lived memory cells. Over time the memory pool itself undergoes a process of functional maturation and conversion of the prominent cell type from effector memory (Tem) to central memory (Tcm) (Obar and Lefrancois, 2010; Wherry et al., 2003b). The role of TCR signaling in these processes has been, for the most part, tested indirectly by manipulating the immunogen or by comparing responses to epitopes that are recognized with different affinities (Kaech and Ahmed, 2001;
weak externally supplied signals diminish the magnitude of initial expansion, hasten T cell contraction following infection and increase the rate of acquisition of Tcm cells. More recently, infection of mice bearing a mutation in the TCRβ chain, which decreases
NFkB activation, provided evidence that distinct TCR signals can direct effector versus memory fates as these mice showed intact effector differentiation but severely defective memory CD8 T cell formation following Listeria monocytogenes (LM) infection (Teixeiro et al., 2009).
Here we explore the role of TCR signals in the differentiation and function of CD4+ T helper cells and CD8+ effector and memory T cells using Y145F and Y112/128F SLP-76 KI mice. Using in vitro Th polarization assays we establish a role for SLP-76 tyrosine signals in Th2 and Th17 function. Using the Armstrong strain of lymphocytic choriomeningitis virus (LCMV) as a model of CD8+ T cell effector and memory differentiation, we show that SLP-76 mutant CD8+ T cells undergo primary clonal expansion, contraction and memory development, despite striking defects in cytokine effector function. Our findings indicate that TCR signals required to initiate a program of T cell differentiation are quantitatively and/or qualitatively different than those required to elicit cytokine responses upon re-stimulation.
Results
Differential requirements for SLP-76 tyrosine signals in T helper lineage
polarization
To begin to address the effects of altered TCR signals on the quality of immune responses we first sought to determine if the dampened signals in SLP-76 KI T cells could support differentiation into Th subsets in vitro. Naïve CD4 T cells were purified and
cultured on a CD3/CD28 stimulating surface in the presence of: IL12 and anti-IL4 for Th1 skewing; IL4, anti-IL12 and anti-IFNγ for Th2 skewing; TGFβ, IL6, IL23 and anti-IFNγ for
Th17 skewing. Following culture, T cells were restimulated with CD3 and CD28 and analyzed for production of signature Th cytokines (Figure 3.1, A, B, C; top panels). SLP- 76 KI T cells produced IFNγ at similar frequencies and levels to WT T cells following Th1
polarization (Figure 3.1A, top panel). However following Th2 and Th17 polarization many fewer SLP-76 KI T cells produced IL4 and IL17a, respectively, compared to their WT counterparts (Figure 3.1B,C; top panels). Furthermore, the mean fluorescence intensity (MFI) of IL4 or IL17a of the populations of SLP-76 KI T cells that did produce cytokine was less than that of the WT cytokine producers, suggesting that SLP-76 KI cells tend to produce less cytokine on a per cell basis (Figure 3.1B,C; top panels). The lack of cytokine production in Th2- and Th17-skewed SLP-76 KI T cells following TCR restimulation could be attributed to either defects in the ability of SLP-76 KI cells to differentiate into competent Th cells or defects in the ability of competent SLP-76 KI Th cells to elicit TCR-induced recall functions. To distinguish between these two
possibilities we restimulated skewed WT and SLP-76 KI T cells with PMA and Ionomycin in order to by-pass proximal TCR signals and looked for the production of signature Th cytokines. As expected, there was no difference in the frequency of IFNγ producers
between Th1-skewed WT and SLP-76 KI T cells following PMA and Ionomycin treatment (Figure 3.1A; bottom panel). There was also no difference in IL-4 production between Th2-skewed WT and SLP-76 KI T cells, suggesting that the SLP-76 KI T cells can undergo Th2 differentiation and produce IL4 competent cells but they have a defect in the ability to elicit IL4 production following TCR restimulation (Figure 3.1B; bottom
that these cells have a defect in the ability to acquire IL17a competence (Figure 3.1C). Despite the SLP-76 tyrosine hierarchy observed in previous signaling and functional assays (see Chapter 2), there were no significant differences between Y145F KI and Y112/128F KI responses under any of the Th skewing conditions.
The requirement of SLP-76 tyrosines for IL17a competence and production
is independent of their role in T cell thymic development
Since SLP-76 tyrosines plays a role in T cell thymic development (see chapter 2) it is possible that inability of naïve SLP-76 KI T cells to acquire IL17a competence could be due to a developmental defect. We therefore looked at the ability of naïve SLP-76 cKI T cells to acquire and produce IL17a following culture under Th17 polarizing conditions. As previously described in chapter 2, SLP-76 cKI T cells express WT SLP-76 during thymic development but it is later deleted with Tamoxifen-inducible Cre activity, leaving expression of only the mutant form of SLP-76. Following tamoxifen treatment, sorted naïve, YFP+ CD4+ T cells from SLP-76 cKI and cSLP-76+/- mice were cultured under Th17 skewing conditions and then restimulated with either anti-TCR/CD28 or PMA and Ionomycin. Similar to what was observed in non-conditional SLP-76 KI T cells, SLP-76 cKI T cells failed to produce WT levels of IL17a regardless of the restimulating
conditions (Figure 3.4). These data suggest that TCR-induced signals through the SLP- 76 tyrosines are required for naïve CD4 T cell acquisition of IL17a competence
SLP-76 KI Th17 polarized T cells can upregulate the Th17 master
regulator RORγT
We next looked at expression RORγT transcription following Th17 skewing in WT and
SLP-76 KI T cells to determine if SLP-76 KI T cells have a defect in the ability to upregulate this master regulator of Th17 differentiation and Il17a production. The overwhelming majority of both SLP-76 KI and WT T cells expressed RORγT following
Th17 skewing and all cells that produced IL17a following TCR/CD28 or PMA and Ionomycin stimulation expressed RORγt (Figure 3.3). Thus, TCR signals through SLP-
76 tyrosines are required for the acquisition of IL17a competence but not for acquisition of other characteristics of Th17 differentiation including RORγT upregulation, consistent
with the phenotype recently observed in Itk deficient T cells (Gomez-Rodriguez et al., 2009). Future studies are required to determine if SLP-76 KI Th17 cells, like Itk-/- Th17 cells, are competent to produce other cytokines associated with Th17 function including IL17f and IL22.
A subset of splenic CD4
+CD44
hiT cells are poised to produce IL17a
independent of SLP-76 tyrosines
As other T cell types have been shown to have the potential to produce IL17a including iNKT cells, CD8 T cells and γδ T cells (reviewed in (O'Brien et al., 2009)), we next
sought to determine if the defect in the ability to acquire competence to produce IL17a was a global defect amongst T cells in the SLP-76 KI mice. To this end, we stimulated bulk splenocytes from WT and SLP-76 KI mice directly ex vivo with PMA and Ionomycin and looked for the production of IL17a. Among CD5+ WT T cells we found a small
similar populations were found in Y145F KI mice and Y112/128F KI mice and, in fact, the frequency of these cells was greater in the KI mice than in the WT mice (Figure 3.4, top panel). Of the CD5+, IL17a producers almost 30% were found to be γδ T cells in the WT
mice but only around 1% of CD5+, IL17a producers were
γδ T cells in either strain of
SLP-76 KI mice (Figure 3.4, middle panel). There was no difference in the frequency of
γδ T cells in the SLP-76 KI mice when compared to WT mice (data not shown). It is
possible that the CD44hi IL17a producing cells we observed were NKT cells, however, previous reports have shown that IL17a producing NKT cells are CD4- (Michel et al., 2008) and most of the γδ- T cells that produced IL17a were CD4+ in WT and SLP-76 KI
mice (data not shown). In the total CD4+ T cell population, the frequency of IL17a producers was considerably greater in both strains of KI mice when compared to WT (Figure 3.4, bottom panel). The greater frequency of this population of CD4+ T cells in the KI mice may be due to altered thymic selection and studies using cKI mice are underway to test this.
Th17 skewing conditions promote IL17a potential in WT and SLP-76 KI
CD44
hiCD4
+T cells.
We cultured CD62Llo CD44hi CD4+ T cells from WT and SLP-76 KI mice under Th17 polarizing conditions to determine if they could acquire, or further acquire, IL17a potential. Restimulation with anti-TCR/CD28 or PMA and Ionomycin induced almost 10% or 20% of WT cultured cells to produce IL17a, respectively (Figure 3.5, left panels). Surprisingly, cultured T cells from both strains of SLP-76 KI mice produced IL17a not only in response to PMA and Ionomycin but also in response to anti-TCR/CD28. It is still unclear whether culturing conditions promoted the expansion of existing T cells with IL17a potential (which could account for the greater frequency of these cells in the SLP-
76 KI cells cultures) or de novo induction of IL17a T cells or a combination of the two. Taken together these data suggest that unlike induced Th17 cells and γδ T cells, CD44hi
CD4+ T cells can acquire IL17a competence independent of SLP-76 tyrosines and furthermore they can produce IL17a in response to TCR stimulation independent of SLP- 76 tyrosines.
CD8 expansion in response to acute LCMV infection is intact in SLP-76KI
mice
Having established a role for SLP-76 tyrosine signals in T cell differentiation in vitro, we next sought to examine the role of these signals for a complete in vivo immune
response. We chose LCMV infection as it is also a well-studied model of CD8
differentiation. WT, Y145F and Y112/128F mice were infected with the Armstrong strain of LCMV that induces an acute infection in WT mice. Analysis of peripheral blood lymphocytes (PBLs) 5 days before infection (baseline) shows that CD8+ T cells were present at similar frequencies within the blood in the three different strains of mice (Figure 3.6A). In WT mice at day 4 post-infection (p.i.), the percent of CD8+ T cells in the blood was decreased from baseline. It is unclear why this decrease was more dramatic in KI mice (Figure 3.6A). Regardless, by the peak of CD8+ T cell expansion at day 8 p.i. and later at day 15 p.i., the percent of circulating CD8+ T cells was similar between the three strains of mice (Figure3.6A). Absolute numbers of CD8+ T cells were calculated from spleens on day 8 p.i. and confirmed an intact CD8+ T cell expansion in the KI mice (Figure 3.6B). In the Y145F mice, splenic CD8+ T cells were present in higher numbers, although the differences were not significant in all experiments (Figure
proliferative defect in vitro in response to TCR engagement, an in vivo viral challenge is sufficient to induce substantial T cell expansion comparable to that induced in WT mice. We next examined the quality of the T cell response to LCMV in WT and KI mice. Expansion of T cells reactive to specific dominant and subdominant LCMV epitopes were examined using 3 different H2Db: peptide tetramers: NP396, GP33 and GP276 (Figure 3.6C). While expansion of CD8+ T cells reactive to the more dominant epitopes NP396 and GP33 was intact in both strains of KI mice, very few cells reactive to the subdominant epitope, GP276, were observed in the Y145F KI mice (van der Most et al., 1996). Y112/128F KI mice showed a lesser but consistent decrease in the absolute numbers of CD8+ splenocytes reactive to GP276. Thus, expansion in response to particular epitopes may be differentially sensitive to the effectiveness of the TCR signal based on the level of epitope dominance. Alternatively, the decrease of GP276-reactive T cells observed in the SLP-76 KI mice could be due to decreased expansion or
perturbation in selection of cells responsive to this epitope during thymic development or a combination of these possibilities.
Altered effector function upon re-stimulation in SLP-76 KI CD8
+T cells
following LCMV infection
The effector function of CD8+ T cells from day 8 p.i. WT and SLP-76 KI mice was analyzed ex vivo by measuring cytokine production in splenocytes in response to incubation with NP396, GP33 and GP276 peptides. Consistent with tetramer reactivity, fewer SLP-76 KI T cells produced cytokines in response to GP276 peptide compared to WT cells (Figure 3.7A). The amount of IFNγ produced by each cell, as measured by the
mean fluorescence intensity (MFI), was lower in T cells from both strains of KI mice in response to all three peptides (Figure 3.7B). While similar frequencies of CD8+ cells from
the three strains of mice produced IFNγ in response to NP396 and GP33, the ratio of
cells that produced TNFα and IFNγ to cells that produced IFNγ alone was much lower in
the Y145F mice but normal in the Y112/128F mice (Figures 3.7A, C, D). This ratio was decreased in Y112/128F T cells only in response to the subdominant epitope GP276 (Figure 3.7C). Despite the weaker cytokine responses, viral titers from liver and spleen were below the level of detection for all mice by day8 (data not shown).
Conditional SLP-76 KI T cells show defects in their response to LCMV
challenge
To determine if the altered CD8 repertoire and T cell selection observed in the KI mice, especially the Y145F KI mice could account for either the surprising robustness of the SLP-76 KI response or the effector defects in the responding KI T cells, tamoxifen- treated cKI and cSLP-76+/− mice were infected with LCMV Armstrong. At day 8 p.i., the
frequency of YFP+CD8+ T cells reactive to all three LCMV tetramers was not significantly different in either cKI mouse strain, when compared to cSLP-76+/− mice (Figure 3.8A).
This was in contrast to the near absence of GP276 tetramer-reactive T cells observed in Y145F KI mice (Figure 3.7). When challenged with LCMV peptides, cKI T cells behaved